2021
DOI: 10.1016/j.cities.2020.103045
|View full text |Cite
|
Sign up to set email alerts
|

Reading the city through its neighbourhoods: Deep text embeddings of Yelp reviews as a basis for determining similarity and change

Abstract: This paper develops novel methods for using Yelp reviews as a window into the collective representations of a city and its neighbourhoods. Basing analysis on social media data such as Yelp is a challenging task because review data is highly sparse and direct analysis may fail to uncover hidden trends. To this end, we propose a deep autoencoder approach for embedding the language of neighbourhood-based business reviews into a reduced dimensional space that facilitates similarity comparison of neighbourhoods and… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
15
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
8

Relationship

1
7

Authors

Journals

citations
Cited by 18 publications
(15 citation statements)
references
References 51 publications
0
15
0
Order By: Relevance
“…We have demonstrated how qualitative logics of inquiry and skill sets can be folded into computational research. We have demonstrated this by showing some of the iterative steps involved in using computational approaches to examine police communications in ways that cannot be fully automated, and which complement the 'thinner' insights of quantitative methods with the 'thicker' insights of qualitative ones (Bornakke and Due 2018;Jemielniak 2020;Olson et al 2021).…”
Section: Discussionmentioning
confidence: 99%
See 1 more Smart Citation
“…We have demonstrated how qualitative logics of inquiry and skill sets can be folded into computational research. We have demonstrated this by showing some of the iterative steps involved in using computational approaches to examine police communications in ways that cannot be fully automated, and which complement the 'thinner' insights of quantitative methods with the 'thicker' insights of qualitative ones (Bornakke and Due 2018;Jemielniak 2020;Olson et al 2021).…”
Section: Discussionmentioning
confidence: 99%
“…Both quantitative and qualitative research skills and logics are often required when working with computational techniques. Computational approaches to research can help bridge these cultures by productively and creatively drawing from both traditions (Bornakke and Due 2018;Brandt and Timmermans 2021;Jemielniak 2020;Mills 2018;Olson et al 2021). The desirability of small samples for some types of research and some research questions (Benfer 1968;Small 2009) is not incompatible with computational approaches.…”
Section: The Computational Turn In Criminology and Criminal Justice Studiesmentioning
confidence: 99%
“…Zukin, Lindeman and Hurson [49] identify racial bias by manually coding Yelp reviews of restaurants in two gentrifying neighbourhoods -one predominantly Black and another predominantly White. Olson et al [50] use computational text analysis techniques to "read the city" via Yelp reviews, revealing collective representations across neighbourhoods and identifying patterns of continuity and change among them.…”
Section: Yelp As a Source For Studying Neighbourhood Changementioning
confidence: 99%
“…Since then, predictive improvements have been made by capitalizing on machine learning methods and novel, especially user‐generated or microgeographic, data sources. Efforts in “nowcasting” or associating gentrification with Yelp data reviews and business types (Glaeser, Kim, & Luca, 2018; Olson, Calderon‐Figueroa, Bidian, Silver, & Sanner, 2021), Instagram postings (Han, Hong, & Lee, 2020), Foresquare check‐ins (Arribas‐Bel & Bakens, 2019), and restaurant reviews (Dong, Ratti, & Zheng, 2019) have advanced the recent state of the art. The use of more traditional government statistics has also been blended with more sophisticated methods such as random forests to better predict gentrification (Reades, De Souza, & Hubbard, 2019).…”
Section: Looking Forward: Understanding Changes In Near Real Time And...mentioning
confidence: 99%